Cristian Iorga [Thu, 27 Mar 2014 07:18:17 +0000 (09:18 +0200)]
lsb: fix lsb_log_msg() implementation
LSB lsb_log_message calls a begin()
function that should be implemented
in /etc/init.d/functions.
The aforementioned script does not
implement the begin() function, as
such there is a small issue related to logging.
This fix implements a local version of
the function, while cleaning up the
troublesome previous implementation.
Fix [YOCTO #5795]
Signed-off-by: Cristian Iorga <cristian.iorga@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 26 Mar 2014 22:29:20 +0000 (22:29 +0000)]
git: Fix perl paths in scripts and population of the perltools package
References to "perl-native" were slipping into the target packages. These
changes ensure those references are cleaned up and that tools using perl
are packaged in the correct perltools package. The same issues affected
the nativesdk-git output so are also applied there.
[YOCTO #5918]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 26 Mar 2014 15:09:06 +0000 (15:09 +0000)]
sanity.bbclass: Update against bitbake sanity event changes
Bitbake will now trigger sanity events when it needs the checks to run in all cases
so we can drop the ConfigParsed hook. We now control whether events are generated
or errors are raised from the event itself.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 26 Mar 2014 09:30:33 +0000 (09:30 +0000)]
base.bbclass: Run oe_import before other INHERITs
Its possible for classes listed in INHERIT directives to use things like
the oe.utils functions. If that happens the user sees a traceback since
the modules don't become available until the ConfigParsed event.
This change to use immediate expansion means that the oe modules become available
much sooner and can be used in the core classes, including within base.bbclass.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jukka Rissanen [Wed, 26 Mar 2014 14:40:26 +0000 (16:40 +0200)]
meta-skeleton: Add name attribute to SRC_URI
The SRCREV_machine line does not work without having name=machine
attribute in SRC_URI.
This error is seen if the custom kernel recipe is used without
the name attribute:
NOTE: Error during finalise of .../linux-yocto-custom.bb
ERROR: ExpansionError during parsing .../linux-yocto-custom.bb: Failure expanding variable do_patch: ExpansionError: Failure expanding variable SRCPV, expression was ${@bb.fetch2.get_srcrev(d)} which triggered exception FetchError: Fetcher failure for URL: 'git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git;protocol=git;nocheckout=1'. Please set a valid SRCREV for url ['SRCREV_default_pn-linux-yocto-custom', 'SRCREV_default', 'SRCREV_pn-linux-yocto-custom', 'SRCREV'] (possible key names are git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git;protocol=git;nocheckout=1, or use a ;rev=X URL parameter)
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 26 Mar 2014 11:41:31 +0000 (11:41 +0000)]
apt/package_mamager: Ensure WORKDIR is used for lists directory
The native sysroot should not be used as a store for the lists files since
multiple images running at once would conflict over this. Instead redirect
this to WORKDIR. This means some extra directories need to be created.
Also create apt.conf.d to silence some warnings.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 26 Mar 2014 11:40:12 +0000 (11:40 +0000)]
package_manager: Ensure we don't process directories twice
Processing directories twice is both pointless and introduces a race condition.
When building the list, ensure duplicates (like "all" and "noarch") are handled
correctly.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Chase Maupin [Tue, 18 Mar 2014 11:56:13 +0000 (06:56 -0500)]
mmc-utils: Add user space mmc utilities for eMMC
* The mmc-utils are useful userspace utilities for configuring and
working with MMC devices. These are particularly useful when
working with eMMC devices to do the initial programming of the
device.
Signed-off-by: Chase Maupin <Chase.Maupin@ti.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Tue, 25 Mar 2014 02:18:47 +0000 (10:18 +0800)]
meta/conf/bitbake.conf: add STAMPCLEAN to BB_HASHBASE_WHITELIST
The problem is that do_configure.sigdata depends on STAMPS_DIR because:
do_configure -> STAMPCLEAN -> STAMPS_DIR
this will make the sigdata generated by "STAMPS_DIR=/tmp/stps bitbake -S
recipe" doesn't match the ones in our build dir, but it should. We can
add STAMPS_DIR or STAMPCLEAN to BB_HASHBASE_WHITELIST to fix the
problem, but we can't add STAMPS_DIR since once it is in
BB_HASHBASE_WHITELIST, the "STAMPS_DIR=/tmp/stps bitbake -S recipe"
would not run again.
[YOCTO $6031]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Wed, 26 Mar 2014 07:01:29 +0000 (03:01 -0400)]
util-linux-native: fix qsort_r for CentOS 5.10
The qsort_r() was added to glibc in version 2.8, so there is no qsort_r() on
the host like CentOS 5.x, use qsort() to fix it since they are nearly
identical.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jonas Eriksson [Mon, 24 Mar 2014 15:56:05 +0000 (16:56 +0100)]
cpio rootfs build: Avoid modifying rootfs dir
The Linux kernel requires that initrd images contain a /init file for
the image to be used as an initrd, even if it is empty. Adding it into
the rootfs directory creates a race, that can upset tar when building
both a .tar and .cpio image file ("tar: .: file changed as we read it").
Additionally, whether or not the tar file will contain the /init file is
also up to the race condition.
To avoid this problem, move the /init addition out from the rootfs
directory, and thus only include it in the .cpio image.
Signed-off-by: Jonas Eriksson <jonas.eriksson@enea.com> Signed-off-by: Josep Puigdemont <josep.puigdemont@enea.com> Cc: Laurentiu Palcu <laurentiu.palcu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Yong Zhang [Wed, 26 Mar 2014 08:32:13 +0000 (16:32 +0800)]
gnupg: CVE-2013-4576
GnuPG 1.x before 1.4.16 generates RSA keys using sequences of introductions
with certain patterns that introduce a side channel, which allows physically
proximate attackers to extract RSA keys via a chosen-ciphertext attack and
acoustic cryptanalysis during decryption. NOTE: applications are not typically
expected to protect themselves from acoustic side-channel attacks, since this
is arguably the responsibility of the physical device. Accordingly, issues of
this type would not normally receive a CVE identifier. However, for this
issue, the developer has specified a security policy in which GnuPG should
offer side-channel resistance, and developer-specified security-policy
violations are within the scope of CVE.
Ming Liu [Wed, 26 Mar 2014 08:32:12 +0000 (16:32 +0800)]
gnupg: CVE-2013-4351
GnuPG 1.4.x, 2.0.x, and 2.1.x treats a key flags subpacket with all bits
cleared (no usage permitted) as if it has all bits set (all usage permitted),
which might allow remote attackers to bypass intended cryptographic protection
mechanisms by leveraging the subkey.
Signed-off-by: Ming Liu <ming.liu@windriver.com> Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Yue Tao [Wed, 26 Mar 2014 09:08:45 +0000 (17:08 +0800)]
Security Advisory - openssl - CVE-2013-6449
The ssl_get_algorithm2 function in ssl/s3_lib.c in OpenSSL before 1.0.2
obtains a certain version number from an incorrect data structure, which
allows remote attackers to cause a denial of service (daemon crash) via
crafted traffic from a TLS 1.2 client.
Signed-off-by: Yue Tao <Yue.Tao@windriver.com> Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Yue Tao [Wed, 26 Mar 2014 09:08:44 +0000 (17:08 +0800)]
Security Advisory - openssl - CVE-2013-6450
The DTLS retransmission implementation in OpenSSL through 0.9.8y and 1.x
through 1.0.1e does not properly maintain data structures for digest and
encryption contexts, which might allow man-in-the-middle attackers to
trigger the use of a different context by interfering with packet delivery,
related to ssl/d1_both.c and ssl/t1_enc.c.
Signed-off-by: Yue Tao <Yue.Tao@windriver.com> Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Yue Tao [Wed, 26 Mar 2014 09:08:43 +0000 (17:08 +0800)]
Security Advisory - openssl - CVE-2013-4353
The ssl3_take_mac function in ssl/s3_both.c in OpenSSL 1.0.1 before
1.0.1f allows remote TLS servers to cause a denial of service (NULL
pointer dereference and application crash) via a crafted Next Protocol
Negotiation record in a TLS handshake.
Signed-off-by: Yue Tao <Yue.Tao@windriver.com> Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Since yasm has been moved to OE-core, there is no reason for not enabling
yasm by default anymore. It improves performance of gstreamer1.0-libav
considerably.
Signed-off-by: Carlos Rafael Giani <dv@pseudoterminal.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Valentin Popa [Tue, 25 Mar 2014 12:06:08 +0000 (14:06 +0200)]
piglit: ship files with correct permissions
Adress the warning
| warning: group pulse does not exist - using root
Piglit files from framework/ generated_tests/ tests/
and templates/ belong to xuser/pulse.
Don't keep the permissions while shipping the files,
they should be root/root.
[YOCTO #6028]
Signed-off-by: Valentin Popa <valentin.popa@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Tue, 25 Mar 2014 10:29:26 +0000 (10:29 +0000)]
binutils: Add fixes for binutils issue 16428
"ld:i386 crashes with -static -fPIE -pie"
https://bugs.launchpad.net/ubuntu/+source/binutils/+bug/1266492
This adds in two upstream binutils fixes to avoid the internal error
triggered by the combination of -static with -pie on x86 builds. This
triggers a backtrace which then triggers a bug in glibc where the process
ends up hanging on some systems with broken libcs.
We can't fix the libc but we can stop the internal error and hence
avoid the hanging builds.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Tue, 25 Mar 2014 08:34:14 +0000 (16:34 +0800)]
tcl: fix path in Config.sh for sstate
We need fix the path in tclConfig.sh, tdbcConfig.sh and itclConfig.sh
for sstate, otherwise there would be build failures when use the sstate
across different builds.
e.g., when building expect:
[snip]
tmp/sysroots/qemuarma9/usr/include/tcl8.6
checking for Tcl private include files... configure: error: Cannot find private header tclInt.h in
/path/to/another/build/tmp/sysroots/qemuarma9/usr/include/tcl8.6.1
Configure failed.
[snip]
[YOCTO #6035]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Mon, 24 Mar 2014 15:18:52 +0000 (15:18 +0000)]
classes/binconfig: fix indentation in python function
If we don't fix this, the recently added dependency from
do_populate_sysroot on functions in SYSROOT_PREPROCESS_FUNCS in
staging.bbclass triggers warnings that the get_binconfig_mangle
function contains tabs for all recipes that inherit binconfig.
Related to fix for [YOCTO #5852].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Mon, 24 Mar 2014 15:18:51 +0000 (15:18 +0000)]
classes/staging: ensure do_populate_sysroot re-execs on changes to sysroot preprocessing funcs
Without this dependency, changes to functions added to
SYSROOT_PREPROCESS_FUNCS do not change do_populate_sysroot's signature
and thus don't cause it to re-execute.
Fixes [YOCTO #5852].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Marius Avram [Mon, 17 Mar 2014 15:04:10 +0000 (15:04 +0000)]
buildstats: use TaskBase time for elapsed time
To avoid any further inconsistencies between buildstats and buildinfohelper
from toaster, buildstats will measure task duration using the time field
from within the TaskBase events: TaskStarted and TaskSucceeded/TaskFailed.
Signed-off-by: Marius Avram <marius.avram@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Thu, 20 Mar 2014 11:39:20 +0000 (11:39 +0000)]
staging: Save out provider information into the sysroot
This patch saves out provider information into the sysroot for all providers
except those in MULTI_PROVIDER_WHITELIST. This means that we will start seeing
warnings when two providers for the same thing are installed into the sysroot.
In the future those warnings can be turned into errors.
Partially addresses [YOCTO #4102]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Mon, 24 Mar 2014 05:39:27 +0000 (13:39 +0800)]
archiver.bbclass: emit patch series files for original src
Emit patch series files for original src so that the user knows how to
apply the patch orderly.
[YOCTO #5113]
Signed-off-by: Christopher Larson <kergoth@gmail.com> Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Mon, 24 Mar 2014 05:39:26 +0000 (13:39 +0800)]
archiver.bbclass: make it can filter the license
* Filter the license (default: no), the recipe whose license in
COPYLEFT_LICENSE_INCLUDE will be included, and in
COPYLEFT_LICENSE_EXCLUDE will be excluded.
* The user can set the recipe type that would be archived (native,
target, and so on), deafult to all.
The copyleft_filter.bbclass is come from copyleft_compliance.bbclass,
which is used by both copyleft_compliance.bbclass and archiver.bbclass.
[YOCTO #5740]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Fri, 21 Mar 2014 18:21:59 +0000 (18:21 +0000)]
gtk-engines: move engine schemas to a dedicated package
The GTK+ engine schemas are not insignicant in size but also mostly useless.
Put them in a sub-package instead of PN so they only get installed when
explicitly required.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jonathan Liu [Mon, 24 Mar 2014 00:23:02 +0000 (11:23 +1100)]
connman: upgrade to 1.22
Highlights:
- If VPN is split routed and not the default service, then allow
DNS queries also to be sent to VPN DNS server.
- Session API fixes
- Memory leak fixes
- Crash fixes
- NTP kiss-of-death packet support
- Support for full USB gadget networking. Now USB gadget network can
be used without tethering.
Signed-off-by: Jonathan Liu <net147@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Chong Lu [Mon, 24 Mar 2014 08:12:47 +0000 (16:12 +0800)]
acl: fix the order of expected output of getfacl
The result of getfacl is sorted by user id.
In Centos or RHEL, bin user id is 1 and daemon user id is 2.
But in our image, bin user id is 2 and daemon user id is 1.
The patch fixes this issue to make ptest pass.
Signed-off-by: Chong Lu <Chong.Lu@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Fri, 21 Mar 2014 18:02:39 +0000 (18:02 +0000)]
classes/image: ignore modules.* changing during multilib image construction
Since we now run depmod when building images (as the postinst that does
this is now on kernel-base instead of kernel-image) it is possible to
have module file differences between the two halves of the multilib image,
and the code that checks for such differences detects this and fails.
Whitelist this file to avoid the failure.
Specifically, modules.alias, modules.dep and modules.symbol can differ
along with their .bin counterparts.
Related to fix for [YOCTO #5392].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Fri, 21 Mar 2014 18:02:38 +0000 (18:02 +0000)]
classes/kernel: move module postinst commands to kernel-base
Since kernel-base is the package that contains the files that depmod
needs to run, we should be running depmod from the kernel-base
postinstall rather than kernel-image.
Fixes [YOCTO #5392].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Fri, 21 Mar 2014 00:25:07 +0000 (00:25 +0000)]
gtk-engines: removed gnome inheritance
Remove gconf class (via gnome) to fix WARN-QA message regarding
--disable-install-schemas option. Add missing intltool-native build dependency
that previously came through gnome.bbclass.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Fri, 21 Mar 2014 16:31:31 +0000 (16:31 +0000)]
classes/image: disable pam nullok_secure with debug-tweaks
If you want passwordless logins to work with pam enabled, then you can't
have "nullok_secure" enabled on pam_unix entries. Add some
postprocessing to change these to "nullok" when debug-tweaks is in
IMAGE_FEATURES, in order to make passwordless logins with PAM work
again.
Fixes [YOCTO #5973].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Mon, 13 Jan 2014 11:06:00 +0000 (19:06 +0800)]
archiver.bbclass: move a few code to copyleft_compliance.bbclass
Move the code which is only used by copyleft_compliance.bbclass from
archiver.bbclassc, and remove the "inherit archiver" from
copyleft_compliance.bbclass.
The archiver.bbclass is used for archiving various types of sources, but
the copyleft_compliance.bbclass is used for analysing the license, they
don't have much relationships.
[YOCTO #4986]
[YOCTO #5113]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Robert Yang [Sun, 5 Jan 2014 14:58:09 +0000 (22:58 +0800)]
archiver.bbclass: refactor it
The archiver didn't work, and there were a few problems, for example:
1) There was no src_dir.org (or orig), but the diff command still use it
2) There were a few duplicated code
3) It didn't archive the source for the native or gcc
4) The work flow is not very well
5) The "subprocess.call('fakeroot cp xxxx'" should be removed
6) And others ...
So that we have to refactor it, the benefits are:
1) Fix the problems and make it work well.
2) Reduce more than 300 lines
3) Make it easy to use.
Hre are the explanation about the bbclass:
This bbclass is used for creating archive for:
1) original (or unpacked) source: ARCHIVER_MODE[src] = "original"
2) patched source: ARCHIVER_MODE[src] = "patched" (default)
3) configured source: ARCHIVER_MODE[src] = "configured"
4) The patches between do_unpack and do_patch:
ARCHIVER_MODE[diff] = "1"
And you can set the one that you'd like to exclude from the diff:
ARCHIVER_MODE[diff-exclude] ?= ".pc autom4te.cache patches"
5) The environment data, similar to 'bitbake -e recipe':
ARCHIVER_MODE[dumpdata] = "1"
6) The recipe (.bb and .inc): ARCHIVER_MODE[recipe] = "1"
All of the above can be packed into a .src.rpm package: (when PACKAGES
!= "")
ARCHIVER_MODE[srpm] = "1"
[YOCTO #4986]
[YOCTO #5113]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Trevor Woerner [Thu, 20 Mar 2014 21:06:38 +0000 (17:06 -0400)]
core-image-minimal: allow user to set size
Allow the user, in their local configuration, to override the size of the
final image. This is useful when creating an image for (for example) an SD
card and the user wants the image to fill up the card as much as possible.
Signed-off-by: Trevor Woerner <trevor.woerner@linaro.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ricardo Neri [Thu, 20 Mar 2014 19:35:33 +0000 (12:35 -0700)]
runqemu: Add option for custom BIOS directory
Add support to specify a directory for custom BIOS, VGA BIOS and
keymaps as supported by qemu (-L option). Even though this can be
done through qemuparams, having this option provides better user
experience by not having to specify a long and cluttered path along
with other qemuparams that the user might want to specify.
This new options assumes first that the path provided is relative to
OECORE_NATIVE_SYSROOT and will check whether it exists before proceeding.
If not, it will treat the provided path as absolute. This provides
the user flexibility to use BIOS binaries generated inside or outside
the OE build environment.
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jonathan Liu [Wed, 19 Mar 2014 23:05:04 +0000 (10:05 +1100)]
base-files: use /dev/root in /etc/fstab for systemd support
systemd does not recognize "rootfs" in /etc/fstab so the root
filesystem is not checked. As a result, the following message
is logged by journalctl:
systemd-fstab-generator[68]: Checking was requested for "rootfs", but it
is not a device
Changing "rootfs" to "/dev/root" in /etc/fstab allows systemd to
check the root filesystem when the kernel is booted with the root
filesystem mounted read-only.
[YOCTO #5950]
Signed-off-by: Jonathan Liu <net147@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Alexandru DAMIAN [Mon, 17 Mar 2014 15:04:09 +0000 (15:04 +0000)]
sstate: list missing files for toaster
Toaster needs to record the attempts to restore
setscene tasks that don't have a sstate file.
We build a list of tasks for which we can't find an
sstate file, and if we're running under Toaster data
collection, we send it off with a MetadataEvent.
Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
GCC 4.8 includes a new runtime library, libatomic, which supports
atomic operations not supported by hardware or the OS. Build it,
so other packages can link against it, if needed.
Signed-off-by: Cosmin Paraschiv <cosmin.paraschiv@freescale.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Wed, 19 Mar 2014 09:23:53 +0000 (17:23 +0800)]
image_types.bbclass: use 4096 instead of 8192 bytes-per-inode
The image not correctly created if 'ptest-pkgs' is in IMAGE_FEATURES,
this is because there is no free inode left. We can use 4096 instead of
8192 bytes-per-inode to fix the problem, and most of the distributions
us 4096, such as Ubuntu, Suse, Fedora and CentOS.
There are another problems:
* There are error message when there is no free inode left if we run the
mke2fs command manually, but they are not in log.do_rootfs.
* The image generation doesn't stop when error happens because mke2fs
doesn't return failed for this case.
Will fix them in other threads.
[YOCTO #5957]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Laurentiu Palcu [Tue, 18 Mar 2014 11:19:20 +0000 (11:19 +0000)]
buildhistory.bbclass: create proper dependency files for SDK
The old functions were calling the list_installed_packages() wrapper
function that only listed the packages in an image rootfs. Even for
target/host SDK. Also, a python crash was possible if 'bitbake -c
populate_sdk core-image-*' was called without calling 'bitbake
core-image-*' first. That's because the wrapper was always looking into
the image rootfs...
This commit fixes the problem and calls the right wrapper for image/sdk.
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Laurentiu Palcu [Tue, 18 Mar 2014 11:09:56 +0000 (11:09 +0000)]
rootfs.py, sdk.py: adjust/create the wrappers for creating installed packages list
Since we created a new PkgsList object that will deal with listing the
installed packages in a rootfs, use the new class both for images and
SDKs in the wrapper functions.
The old list_installed_packages() wrapper listed only the packages inside
an image rootfs. It didn't deal with target/host SDK rootfs's.
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Laurentiu Palcu [Tue, 18 Mar 2014 11:03:51 +0000 (11:03 +0000)]
package_manager.py: create separate class for installed packages listing
This commit creates a new class that has the only purpose to generate
various listings of installed packages in the rootfs.
Basically, the methods involved in listing the installed packages, that
were part of each backend PM class implementation, were moved to this
new class.
This change avoids instantiating a new PM object just to get the list of
installed packages in a certain rootfs.
Signed-off-by: Laurentiu Palcu <laurentiu.palcu@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Wed, 12 Mar 2014 20:30:40 +0000 (20:30 +0000)]
perf: add libexec and traceevent to packages
Recent versions of perf may install files into /usr/libexec/perf-core and in
/usr/lib/traceevent. To avoid packaging QA errors, we add these two
directories to the FILES variables.
We also add: INHIBIT_PACKAGE_DEBUG_SPLIT="1" to avoid the following issue
(due to a trailing / being removed):
ERROR: debugedit failed with exit code 256
...
debugedit: canonicalization unexpectedly shrank by one character
And finally, we must ensure that the traceevent libraries are installed to
the proper multilib library path. If building some multlibs, the incorrect
library path will be selected by perf, since it triggers via: ifeq
($(ARCH),x86_64) (or similiar mechanism per arch).
In a 32 bit build, with a 64 bit multilib, the arch won't match and the
detection of a 64 bit build (and library) are not exected. To ensure that
libraries are installed to the correct location, we can make the substitution
in the config/Makefile. For non multilib builds, this has no impact.
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In an image recipe, you can get a warning, for example:
WARNING: The postinstall intercept hook 'update_font_cache' failed (exit code: 1)! See log for details!
WARNING: The postinstalls for the following packages will be postponed for first boot: ttf-dejavu-sans-mono
(because /usr/bin/fc-cache is missing)
In OE-core, rdepend is correctly done in each recipe:
- ttf-fonts/liberation-fonts
- ttf-fonts/ttf-bitstream-vera
In meta-OE, rdepend is not done but packagegroup-fonts-truetype.bb includes fontconfig-utils:
- ttf-fonts/ttf-arphic-uming
- ttf-fonts/ttf-dejavu
- ttf-fonts/ttf-droid
- ttf-fonts/ttf-gentium
- ttf-fonts/ttf-hunkyfonts
- ttf-fonts/ttf-inconsolata
- ttf-fonts/ttf-liberation
- ttf-fonts/ttf-mplus
- ttf-fonts/ttf-sazanami
- ttf-fonts/ttf-ubuntu-font-family
- ttf-fonts/ttf-wqy-zenhei
Signed-off-by: Matthieu Crapet <Matthieu.Crapet@ingenico.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Thu, 20 Mar 2014 19:24:33 +0000 (19:24 +0000)]
python: add python-mmap to python-multiprocessing RDEPENDS
The multiprocessing module needs the mmap module. We got away with not
having this for the build appliance because python-mmap was added to
packagegroup-self-hosted, but this is the proper place to have it.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Hatle [Thu, 20 Mar 2014 19:19:02 +0000 (14:19 -0500)]
useradd{-static}: Ignore useradds on nativesdk
The code was supposed to ignore both native and nativesdk operations when
using the useradd and useradd-static code. However, somewhere along the way
the code was dropped. This didn't cause any issues until someone enabled the
enforcing mode in the new useradd-static and various nativesdk packages
started to fail.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Tue, 18 Mar 2014 07:54:34 +0000 (15:54 +0800)]
sgml-common-native: fix the generation of sgml-docbook.cat
There would be build errors if multiple builds use the same SSTATE_DIR,
and the builds use the different versions' docbook-sgml-dtd-native, the
problems are: we have multiple versions docbook-sgml-dtd-native:
And they depend on sgml-common-native which installs the file
sysroot/etc/sgml/sgml-docbook.cat, this file is also included in the
sstate cache file (.tar.gz), but both the 3 versions
docbook-sgml-dtd-native may update the
sysroot/ect/sgml/sgml-docbook.cat, and it is a hardlink to
SYSROOT_DESTDIR/etc/sgml/sgml-docbook.cat, so the sgml-docbook.cat in
the sstate cache file may contain the
docbook-sgml-dtd-<version>-native's info, and there would be errors when
we mirror the sstate-cache to another build which uses a different
version.
Now we exclude the sgml-docbook.cat from the ${D}, and generate
sysroot/ect/sgml/sgml-docbook.cat dynamically will fix the problem, both
the sgml-common-native and docbook-sgml-dtd-<version>-native can update
it it correctly.
The similar to docbook-dsssl-stylesheets-native and openjade-native.
[YOCTO #5994]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Chen Qi [Tue, 18 Mar 2014 07:12:08 +0000 (15:12 +0800)]
expect: fix do_install failure on SLED 11.2
Previously we were encountering do_install failures on older machines
like SLED 11.2. These machines have relatively lower versions of glibc.
tclsh8.6: /lib64/libc.so.6: version `GLIBC_2.14' not found
The above failure is due to that fact that we are using the fixline1
tcl script to install commands under the example directory. The purpose
of fixline1 is to ensure that the installed scripts begin with '#!/bin/sh'.
However, the scripts under the example directory have already got
the correct first line. That's why we don't need to modify anything and could
just copy them.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
package_manager: Use apt-ftparchive for deb packages
Use apt-ftparchive to create a Release file compatible with SecureApt.
apt-ftparchive is also a more efficient replacement of
dpkg-scanpackages:
root@neopili:~/curro/qtec/qt5022/build-qt5022-cesium/build/tmp/deploy/deb/bobcat
_64# time PSEUDO_UNLOAD=1 apt-ftparchive packages . >/tmp/kkk
real 0m26.873s
user 0m20.968s
sys 0m1.212s
root@neopili:~/curro/qtec/qt5022/build-qt5022-cesium/build/tmp/deploy/deb/bobcat
_64# time PSEUDO_UNLOAD=1 dpkg-scanpackages . >/tmp/kkk
dpkg-scanpackages: info: Wrote 6022 entries to output Packages file.
real 0m59.721s
user 0m16.668s
sys 0m11.164s
apt-ftparchive is not compatible with libpseudo. The calls to ftw()
returns the path in absolute format instead of relative. This
produces
wrong Packages and Release files.
gettext-0.16.1: kill target m4 macros from sysroot
Ever since the change to how aclocal files are copied (based on dependencies),
target m4 macros seem to more reliably be used in preference to native (which
they should), but in a non-gplv3 build, gettext is 0.16 while gettext-native is
0.18, causing a 0.16 po.m4 to be used with our 0.18 po/Makefile.in.in files,
causing at least some failed builds, including e2fsprogs.
Anyone inheriting gettext will have both gettext-native and gettext available,
and we don't want to use older macros from the target gettext in a non-gplv3
build, so kill them and let dependent recipes rely on gettext-native.
[YOCTO #5964]
Signed-off-by: Christopher Larson <kergoth@gmail.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
